Systems Frontend issueshttps://gitlab.rlp.net/motionbank/applications/systems-frontend/-/issues2024-03-11T12:51:57+01:00https://gitlab.rlp.net/motionbank/applications/systems-frontend/-/issues/2222Filter Term Frequency Analysis is broken2024-03-11T12:51:57+01:00Hansen, ChristianFilter Term Frequency Analysis is broken![Bildschirmfoto_2024-03-11_um_12.49.45](/uploads/1b06a230d83153cfbd57e2f1c292b603/Bildschirmfoto_2024-03-11_um_12.49.45.png)![Bildschirmfoto_2024-03-11_um_12.49.45](/uploads/1b06a230d83153cfbd57e2f1c292b603/Bildschirmfoto_2024-03-11_um_12.49.45.png)https://gitlab.rlp.net/motionbank/applications/systems-frontend/-/issues/22213d feature improvement2024-02-02T13:17:13+01:00Hansen, Christian3d feature improvementThis is the related branch:
<br />
https://gitlab.rlp.net/motionbank/applications/systems-frontend/-/tree/three-d-feature-extending?ref_type=headsThis is the related branch:
<br />
https://gitlab.rlp.net/motionbank/applications/systems-frontend/-/tree/three-d-feature-extending?ref_type=headsHansen, ChristianHansen, Christianhttps://gitlab.rlp.net/motionbank/applications/systems-frontend/-/issues/2210magic menu: "3d" button is missing suddenly2024-01-09T12:48:48+01:00Hansen, Christianmagic menu: "3d" button is missing suddenly![Bildschirmfoto_2023-12-19_um_17.35.39](/uploads/c202da6d87b82be0adc65c6bea0b4ed8/Bildschirmfoto_2023-12-19_um_17.35.39.png)![Bildschirmfoto_2023-12-19_um_17.35.39](/uploads/c202da6d87b82be0adc65c6bea0b4ed8/Bildschirmfoto_2023-12-19_um_17.35.39.png)https://gitlab.rlp.net/motionbank/applications/systems-frontend/-/issues/2209post annotator: automatic transcription after video file (not audio file) upl...2024-01-10T17:41:38+01:00Hansen, Christianpost annotator: automatic transcription after video file (not audio file) upload in post-annotator-input is missing![Bildschirmfoto_2023-12-19_um_12.14.21](/uploads/a45b020040dfa35a6825de54baada53e/Bildschirmfoto_2023-12-19_um_12.14.21.png)![Bildschirmfoto_2023-12-19_um_12.14.21](/uploads/a45b020040dfa35a6825de54baada53e/Bildschirmfoto_2023-12-19_um_12.14.21.png)https://gitlab.rlp.net/motionbank/applications/systems-frontend/-/issues/2201magic menu: stop 'continue polling Jobs' interval when leaving post annotator2024-01-09T15:21:32+01:00Hansen, Christianmagic menu: stop 'continue polling Jobs' interval when leaving post annotatorIt unnecessarily spams the whole console:
<br /><br />![Bildschirmfoto_2023-12-07_um_16.29.24](/uploads/32cec3bad144f07946a35d354adb98a1/Bildschirmfoto_2023-12-07_um_16.29.24.png)It unnecessarily spams the whole console:
<br /><br />![Bildschirmfoto_2023-12-07_um_16.29.24](/uploads/32cec3bad144f07946a35d354adb98a1/Bildschirmfoto_2023-12-07_um_16.29.24.png)https://gitlab.rlp.net/motionbank/applications/systems-frontend/-/issues/2200input field -> voice input: add "running process" indicator (spinner) after s...2023-12-13T14:53:08+01:00Hansen, Christianinput field -> voice input: add "running process" indicator (spinner) after stopping a record until record is readySometimes between record stop and having access to this record in the input field (= submit/clear buttons appear) some seconds go by without any indicator, that a converting process is running in the background. This confused me several ...Sometimes between record stop and having access to this record in the input field (= submit/clear buttons appear) some seconds go by without any indicator, that a converting process is running in the background. This confused me several time, because I thought the record broke in some way. That's why we need an "something is happening" indicator as done in the magic menu.https://gitlab.rlp.net/motionbank/applications/systems-frontend/-/issues/2197after rewrite: replace icon for "speech to text"2023-11-28T14:46:17+01:00Hansen, Christianafter rewrite: replace icon for "speech to text"In quasar v1.x some google icons aren't available, so the "speech to text" icon. Should be replaced after updating.
![Bildschirmfoto_2023-11-28_um_14.19.12](/uploads/aff19e232a48c97a5688ebb002722848/Bildschirmfoto_2023-11-28_um_14.19.12....In quasar v1.x some google icons aren't available, so the "speech to text" icon. Should be replaced after updating.
![Bildschirmfoto_2023-11-28_um_14.19.12](/uploads/aff19e232a48c97a5688ebb002722848/Bildschirmfoto_2023-11-28_um_14.19.12.png)
![Bildschirmfoto_2023-11-28_um_14.18.57](/uploads/b6ca64be3dd9ff1d3e9a7ad74a3eff49/Bildschirmfoto_2023-11-28_um_14.18.57.png)https://gitlab.rlp.net/motionbank/applications/systems-frontend/-/issues/2188transcription strings have double spaces between words2023-11-13T15:26:16+01:00Hansen, Christiantranscription strings have double spaces between words![Bildschirmfoto_2023-11-10_um_16.06.53](/uploads/cd57c546e1789140825e94cfc338b16d/Bildschirmfoto_2023-11-10_um_16.06.53.png)
<br /><br />
use single spaces instead![Bildschirmfoto_2023-11-10_um_16.06.53](/uploads/cd57c546e1789140825e94cfc338b16d/Bildschirmfoto_2023-11-10_um_16.06.53.png)
<br /><br />
use single spaces insteadhttps://gitlab.rlp.net/motionbank/applications/systems-frontend/-/issues/2186Add segementation mode to post annotator2023-11-08T09:42:12+01:00Rittershaus, DavidAdd segementation mode to post annotatorThis came up when Helena was using the post annotator for the "inter-rater reliability test" where she wanted to label periodic segements of the video of the exact same length. There was also some feedback of more "pro" users from the di...This came up when Helena was using the post annotator for the "inter-rater reliability test" where she wanted to label periodic segements of the video of the exact same length. There was also some feedback of more "pro" users from the digital humanities field who were wondering why annotations do not have a length by default.
From what I can tell, two different ways of implementing better options for rigid segementation could be helpful:
1) Option to segment the video in a predefined interval (e.g. 1 second). The swimlane then offers "empty" bars. If you select one and add an annotations via input field, you add an annotation which has the start and end time of the selected segment. Visually I'm thinking about something similar to the kinesphere visualisation in the swimlane.
2) Option to switch to a way of annotating that is still more "free". However, you will add annotations with duration by default. The duration is always from the start time to the start time of the next annotation or to the end of the video if there is no other annotation behind the one you add.
Not sure how to best implement this without introducing too much complexity to the UI. However, since these are more "pro" features, they might be a little bit more hidden.
@maximilian.mittermei: do you think that this is more for the rewirte? There is not really a use case within the vortanz project at the moment, so the priority is not very high. However, it would be great to test and improve it.https://gitlab.rlp.net/motionbank/applications/systems-frontend/-/issues/2180Abstraction of skeleton in series2023-10-25T16:41:19+02:00Hansen, ChristianAbstraction of skeleton in seriesAdd optional, customizable skeleton abstraction feature (as seen in the original Mr. Griddle) to series player.Add optional, customizable skeleton abstraction feature (as seen in the original Mr. Griddle) to series player.https://gitlab.rlp.net/motionbank/applications/systems-frontend/-/issues/2176add loading-skeleton for videos2023-10-11T12:00:53+02:00Hansen, Christianadd loading-skeleton for videos![Bildschirmfoto_2023-10-11_um_11.59.01](/uploads/084c06fbfbb837c6c412bde222ecc4c6/Bildschirmfoto_2023-10-11_um_11.59.01.png)![Bildschirmfoto_2023-10-11_um_11.59.01](/uploads/084c06fbfbb837c6c412bde222ecc4c6/Bildschirmfoto_2023-10-11_um_11.59.01.png)https://gitlab.rlp.net/motionbank/applications/systems-frontend/-/issues/2174Bounding boxes: add possibility to select more than one bounding box at once?2023-10-25T16:53:56+02:00Hansen, ChristianBounding boxes: add possibility to select more than one bounding box at once?Example "Effect": "Person A and Person B are looking at each other"Example "Effect": "Person A and Person B are looking at each other"https://gitlab.rlp.net/motionbank/applications/systems-frontend/-/issues/2113post annotator: breakpoints need revision2023-08-10T15:35:51+02:00Hansen, Christianpost annotator: breakpoints need revision![Bildschirmaufnahme_2023-08-10_um_15.34.30](/uploads/e29d9f669b967708f5444856d9434398/Bildschirmaufnahme_2023-08-10_um_15.34.30.mov)![Bildschirmaufnahme_2023-08-10_um_15.34.30](/uploads/e29d9f669b967708f5444856d9434398/Bildschirmaufnahme_2023-08-10_um_15.34.30.mov)https://gitlab.rlp.net/motionbank/applications/systems-frontend/-/issues/2112overlapping videos in post annotator: adjust preview position to current time...2023-08-04T17:26:06+02:00Hansen, Christianoverlapping videos in post annotator: adjust preview position to current time from selected videoYes, this would need some rebuild (using video players instead of preview videos inside the overlay), and maybe it could cause some performance issues. But I think this could be a nice, little detail to have a better understanding how th...Yes, this would need some rebuild (using video players instead of preview videos inside the overlay), and maybe it could cause some performance issues. But I think this could be a nice, little detail to have a better understanding how this whole overlay thing is meant:
<br /><br />
![Bildschirmfoto_2023-08-04_um_17.18.28](/uploads/abc551ba7a975b2aa45b360118db402d/Bildschirmfoto_2023-08-04_um_17.18.28.png)https://gitlab.rlp.net/motionbank/applications/systems-frontend/-/issues/2077Improve visibility for lane labels/titles in swimlane2023-07-03T15:23:55+02:00Rittershaus, DavidImprove visibility for lane labels/titles in swimlaneThis is a minor thing but was part of the feedback from Cologne. I don't see a high priority but it might be relevant in case a "custom lane" option is added (#2076) to make sure you can read the filter keyword that has been applied to a...This is a minor thing but was part of the feedback from Cologne. I don't see a high priority but it might be relevant in case a "custom lane" option is added (#2076) to make sure you can read the filter keyword that has been applied to a lane. It could be addressed at the same time.
Main issue is that names/labels/titles are hidden by annotations or bar graphs:
![Bildschirmfoto_2023-06-28_um_16.59.10](/uploads/479ee5833c9d5bc0e21d0580f8fcb528/Bildschirmfoto_2023-06-28_um_16.59.10.png)
No idea yet how to solve it nicely because I think the labels should not take too much space within the siwmlane (at the moment I'm quite happy with how much information is condensed into a rather thin/small UI element).Mittermeier, MaximilianMittermeier, Maximilianhttps://gitlab.rlp.net/motionbank/applications/systems-frontend/-/issues/2076Add "custom lane" option to swimlane2023-07-03T15:04:41+02:00Rittershaus, DavidAdd "custom lane" option to swimlane@maximilian.mittermei: this could be a small "on the side" project since you mentioned a use case for yourself / your project. The simple solution that I have in mind: there is an "add custom lane" option in the siwmlane menu that brings...@maximilian.mittermei: this could be a small "on the side" project since you mentioned a use case for yourself / your project. The simple solution that I have in mind: there is an "add custom lane" option in the siwmlane menu that brings up a small overlay where you can select filters like in the filter menu for the annotation list. You can then use hashtags like "#lightcue" or "#sound" in the annotations and add seperate lanes for each hashtag. Symbol (image, video, audio) or keyword could be displayed in the lane (as some sort of title) to see what each lane is about.
I guess this one is relevant to not loose all custom lanes right away if you reload the page: #2068Mittermeier, MaximilianMittermeier, Maximilianhttps://gitlab.rlp.net/motionbank/applications/systems-frontend/-/issues/2045Comparison overlay: accessibility check after window resize2023-05-10T16:03:02+02:00Hansen, ChristianComparison overlay: accessibility check after window resize"If the window is resized, there is no check to make sure the popup stays within the window and is reachable:
![Screen_Recording_2023-05-10_at_14.40.53](/uploads/e73312fac415b7c9eba98c8115a6d554/Screen_Recording_2023-05-10_at_14.40.53.m..."If the window is resized, there is no check to make sure the popup stays within the window and is reachable:
![Screen_Recording_2023-05-10_at_14.40.53](/uploads/e73312fac415b7c9eba98c8115a6d554/Screen_Recording_2023-05-10_at_14.40.53.mov)
~~Also the aspect ratio fitting of the compared with video in the popup seems to use the window dimensions / proportions instead of the available space in the popup but I consider that rather cherry-on-top in terms of priority.~~"https://gitlab.rlp.net/motionbank/applications/systems-frontend/-/issues/2044comparison overlay: adapt "no data" on skeleton to "video starts in" on video2023-05-09T20:22:08+02:00Hansen, Christiancomparison overlay: adapt "no data" on skeleton to "video starts in" on video![Bildschirmfoto_2023-05-09_um_20.21.37](/uploads/5bef44e8e87d452018837565213fb3e2/Bildschirmfoto_2023-05-09_um_20.21.37.png)![Bildschirmfoto_2023-05-09_um_20.21.37](/uploads/5bef44e8e87d452018837565213fb3e2/Bildschirmfoto_2023-05-09_um_20.21.37.png)https://gitlab.rlp.net/motionbank/applications/systems-frontend/-/issues/2039Remove "people tracking" from magic menu?2023-05-05T17:49:22+02:00Rittershaus, DavidRemove "people tracking" from magic menu?This came up before but it still seems to be a little bit confusing why there is an option that does not offer anything new compared to "people detection". I know that the data in the background has the IDs when the tracking has be proce...This came up before but it still seems to be a little bit confusing why there is an option that does not offer anything new compared to "people detection". I know that the data in the background has the IDs when the tracking has be processed but this has no effect for users in the vortanz version. My suggestion is to remove it from the menu and to just run it in the background when pose recognition etc. is requested and only use the IDs for the bboxes if they were created through further processing (via pose recognition, kinesphere etc.).https://gitlab.rlp.net/motionbank/applications/systems-frontend/-/issues/2029Video control bar: Open tooltips downwards2023-05-02T16:18:01+02:00Hansen, ChristianVideo control bar: Open tooltips downwardsOtherwise they could cover parts of the video:
<br />
![Bildschirmfoto_2023-04-25_um_15.39.24](/uploads/2b6f7be5ee3f7c961b24516073d1c1ae/Bildschirmfoto_2023-04-25_um_15.39.24.png)Otherwise they could cover parts of the video:
<br />
![Bildschirmfoto_2023-04-25_um_15.39.24](/uploads/2b6f7be5ee3f7c961b24516073d1c1ae/Bildschirmfoto_2023-04-25_um_15.39.24.png)